Unverified50% confidenceSolutionExact time
If a multilingual model's vocabulary has abnormally high fertility on low-resource languages, the problem may not lie in vocabulary size but in the construction objective.
1
Sources
50%
Confidence
Long-term
Relevance
10/13/2026
First Seen
Sources
Related Entities
Related Claims
UnverifiedTokenizers lead to limited vocabularies, uneven support for multilingual and rare characters, and shortcomings in spelling and character-level tasks.75% similarVerifiedText generated by large language models tends to have low perplexity, with word choices that are overly consistent with probabilistic expectations, lacking the unexpectedness of human writing.74% similarVerifiedLarge language models exhibit emergent capabilities, suddenly displaying abilities not present at smaller scales once the model size exceeds a certain threshold.74% similarVerified大语言模型推理的瓶颈往往不在计算能力，而在于内存带宽74% similarUnverifiedThe core limitations of small language models are insufficient knowledge density, weak generalization capability, and an extremely high degree of dependence on prompts.74% similar
Cite This Claim
Stable URI
https://kongchang.com/claim/1013352API
curl https://kongchang.com/api/v1/knowledge/claims/1013352MCP
get_claim(id=1013352)